scholarly journals Symmetric Metric Learning with Adaptive Margin for Recommendation

2020 ◽  
Vol 34 (04) ◽  
pp. 4634-4641
Author(s):  
Mingming Li ◽  
Shuai Zhang ◽  
Fuqing Zhu ◽  
Wanhui Qian ◽  
Liangjun Zang ◽  
...  
Keyword(s):  
Recommender Systems ◽  
Metric Space ◽  
State Of The Art ◽  
Metric Learning ◽  
Current User ◽  
Experimental Results ◽  
Negative Item ◽  
Positive Item ◽  
User Centric ◽  
The Impact

Metric learning based methods have attracted extensive interests in recommender systems. Current methods take the user-centric way in metric space to ensure the distance between user and negative item to be larger than that between the current user and positive item by a fixed margin. While they ignore the relations among positive item and negative item. As a result, these two items might be positioned closely, leading to incorrect results. Meanwhile, different users usually have different preferences, the fixed margin used in those methods can not be adaptive to various user biases, and thus decreases the performance as well. To address these two problems, a novel Symmetic Metric Learning with adaptive margin (SML) is proposed. In addition to the current user-centric metric, it symmetically introduces a positive item-centric metric which maintains closer distance from positive items to user, and push the negative items away from the positive items at the same time. Moreover, the dynamically adaptive margins are well trained to mitigate the impact of bias. Experimental results on three public recommendation datasets demonstrate that SML produces a competitive performance compared with several state-of-the-art methods.

scholarly journals Evaluation of recent advances in recommender systems on Arabic content

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Mehdi Srifi ◽  
Ahmed Oussous ◽  
Ayoub Ait Lahcen ◽  
Salma Mouline
Keyword(s):  
Recommender Systems ◽  
High Performance ◽  
Large Scale ◽  
State Of The Art ◽  
Experimental Results ◽  
Recent Advances ◽  
Research Gap ◽  
Text Preprocessing

AbstractVarious recommender systems (RSs) have been developed over recent years, and many of them have concentrated on English content. Thus, the majority of RSs from the literature were compared on English content. However, the research investigations about RSs when using contents in other languages such as Arabic are minimal. The researchers still neglect the field of Arabic RSs. Therefore, we aim through this study to fill this research gap by leveraging the benefit of recent advances in the English RSs field. Our main goal is to investigate recent RSs in an Arabic context. For that, we firstly selected five state-of-the-art RSs devoted originally to English content, and then we empirically evaluated their performance on Arabic content. As a result of this work, we first build four publicly available large-scale Arabic datasets for recommendation purposes. Second, various text preprocessing techniques have been provided for preparing the constructed datasets. Third, our investigation derived well-argued conclusions about the usage of modern RSs in the Arabic context. The experimental results proved that these systems ensure high performance when applied to Arabic content.

scholarly journals Few-Shot Partial-Label Learning

2021 ◽  
Author(s):  
Yunfeng Zhao ◽  
Guoxian Yu ◽  
Lei Liu ◽  
Zhongmin Yan ◽  
Lizhen Cui ◽  
...  
Keyword(s):  
State Of The Art ◽  
Learning Algorithms ◽  
Metric Learning ◽  
Experimental Results ◽  
Superior Performance ◽  
Distance Metric Learning ◽  
Distance Metric ◽  
Support Set ◽  
Partial Label Learning ◽  
Noise Tolerant

Partial-label learning (PLL) generally focuses on inducing a noise-tolerant multi-class classifier by training on overly-annotated samples, each of which is annotated with a set of labels, but only one is the valid label. A basic promise of existing PLL solutions is that there are sufficient partial-label (PL) samples for training. However, it is more common than not to have just few PL samples at hand when dealing with new tasks. Furthermore, existing few-shot learning algorithms assume precise labels of the support set; as such, irrelevant labels may seriously mislead the meta-learner and thus lead to a compromised performance. How to enable PLL under a few-shot learning setting is an important problem, but not yet well studied. In this paper, we introduce an approach called FsPLL (Few-shot PLL). FsPLL first performs adaptive distance metric learning by an embedding network and rectifying prototypes on the tasks previously encountered. Next, it calculates the prototype of each class of a new task in the embedding network. An unseen example can then be classified via its distance to each prototype. Experimental results on widely-used few-shot datasets demonstrate that our FsPLL can achieve a superior performance than the state-of-the-art methods, and it needs fewer samples for quickly adapting to new tasks.

The Impact of Cash Mobs in the Vote with the Wallet Game: Experimental Results

2016 ◽  
Author(s):  
Leonardo Becchetti ◽  
Maurizio Fiaschetti ◽  
Francesco Salustri
Keyword(s):  
Experimental Results ◽  
The Impact

Browser Security Attacks and Detection Techniques: A Case of Tabnabbing

2020 ◽  
Vol 8 (1) ◽  
pp. 33-41
Author(s):  
Dr. S. Sarika ◽  
Keyword(s):  
Credit Card ◽  
State Of The Art ◽  
Experimental Results ◽  
Detection Technique ◽  
Security Attacks ◽  
Agent Based ◽  
Detection Techniques ◽  
Browser Security ◽  
Cyber Threats ◽  
Multi Agent

Phishing is a malicious and deliberate act of sending counterfeit messages or mimicking a webpage. The goal is either to steal sensitive credentials like login information and credit card details or to install malware on a victim’s machine. Browser-based cyber threats have become one of the biggest concerns in networked architectures. The most prolific form of browser attack is tabnabbing which happens in inactive browser tabs. In a tabnabbing attack, a fake page disguises itself as a genuine page to steal data. This paper presents a multi agent based tabnabbing detection technique. The method detects heuristic changes in a webpage when a tabnabbing attack happens and give a warning to the user. Experimental results show that the method performs better when compared with state of the art tabnabbing detection techniques.

scholarly journals Experimental study of the supercritical CO 2 diffusion coefficient in porous media under reservoir conditions

2019 ◽  
Vol 6 (6) ◽  
pp. 181902 ◽  
Author(s):  
Junchen Lv ◽  
Yuan Chi ◽  
Changzhong Zhao ◽  
Yi Zhang ◽  
Hailin Mu
Keyword(s):  
Porous Media ◽  
Diffusion Coefficient ◽  
Oil Recovery ◽  
Carbon Capture ◽  
Carbon Capture And Storage ◽  
Elevated Pressure ◽  
Experimental Results ◽  
Saturated Porous Media ◽  
Reservoir Conditions ◽  
The Impact

Reliable measurement of the CO 2 diffusion coefficient in consolidated oil-saturated porous media is critical for the design and performance of CO 2 -enhanced oil recovery (EOR) and carbon capture and storage (CCS) projects. A thorough experimental investigation of the supercritical CO 2 diffusion in n -decane-saturated Berea cores with permeabilities of 50 and 100 mD was conducted in this study at elevated pressure (10–25 MPa) and temperature (333.15–373.15 K), which simulated actual reservoir conditions. The supercritical CO 2 diffusion coefficients in the Berea cores were calculated by a model appropriate for diffusion in porous media based on Fick's Law. The results show that the supercritical CO 2 diffusion coefficient increases as the pressure, temperature and permeability increase. The supercritical CO 2 diffusion coefficient first increases slowly at 10 MPa and then grows significantly with increasing pressure. The impact of the pressure decreases at elevated temperature. The effect of permeability remains steady despite the temperature change during the experiments. The effect of gas state and porous media on the supercritical CO 2 diffusion coefficient was further discussed by comparing the results of this study with previous study. Based on the experimental results, an empirical correlation for supercritical CO 2 diffusion coefficient in n -decane-saturated porous media was developed. The experimental results contribute to the study of supercritical CO 2 diffusion in compact porous media.

scholarly journals Comparison of Industrial Quenching Oils

Metals ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 250
Author(s):  
Jiří Hájek ◽  
Zaneta Dlouha ◽  
Vojtěch Průcha
Keyword(s):  
State Of The Art ◽  
Cooling Capacity ◽  
Steel Grade ◽  
Cooling Curves ◽  
Cooling Process ◽  
Industrial Practice ◽  
Product Size ◽  
Cct Curve ◽  
Hardening Power ◽  
The Impact

This article is a response to the state of the art in monitoring the cooling capacity of quenching oils in industrial practice. Very often, a hardening shop requires a report with data on the cooling process for a particular quenching oil. However, the interpretation of the data can be rather difficult. The main goal of our work was to compare various criteria used for evaluating quenching oils. Those of which prove essential for operation in tempering plants would then be introduced into practice. Furthermore, the article describes monitoring the changes in the properties of a quenching oil used in a hardening shop, the effects of quenching oil temperature on its cooling capacity and the impact of the water content on certain cooling parameters of selected oils. Cooling curves were measured (including cooling rates and the time to reach relevant temperatures) according to ISO 9950. The hardening power of the oil and the area below the cooling rate curve as a function of temperature (amount of heat removed in the nose region of the Continuous cooling transformation - CCT curve) were calculated. V-values based on the work of Tamura, reflecting the steel type and its CCT curve, were calculated as well. All the data were compared against the hardness and microstructure on a section through a cylinder made of EN C35 steel cooled in the particular oil. Based on the results, criteria are recommended for assessing the suitability of a quenching oil for a specific steel grade and product size. The quenching oils used in the experiment were Houghto Quench C120, Paramo TK 22, Paramo TK 46, CS Noro MO 46 and Durixol W72.

scholarly journals Smoothing additive manufactured parts using ns-pulsed laser radiation

2021 ◽  
Author(s):  
Florian Kuisat ◽  
Fernando Lasagni ◽  
Andrés Fabián Lasagni
Keyword(s):  
Surface Roughness ◽  
Mechanical Performance ◽  
State Of The Art ◽  
Pulsed Laser ◽  
Industrial Applications ◽  
Laser Source ◽  
Pulsed Laser Radiation ◽  
Current State ◽  
The Impact

AbstractIt is well known that the surface topography of a part can affect its mechanical performance, which is typical in additive manufacturing. In this context, we report about the surface modification of additive manufactured components made of Titanium 64 (Ti64) and Scalmalloy®, using a pulsed laser, with the aim of reducing their surface roughness. In our experiments, a nanosecond-pulsed infrared laser source with variable pulse durations between 8 and 200 ns was applied. The impact of varying a large number of parameters on the surface quality of the smoothed areas was investigated. The results demonstrated a reduction of surface roughness Sa by more than 80% for Titanium 64 and by 65% for Scalmalloy® samples. This allows to extend the applicability of additive manufactured components beyond the current state of the art and break new ground for the application in various industrial applications such as in aerospace.

scholarly journals Automatic Detection of Discrimination Actions from Social Images

Electronics ◽  
2021 ◽  
Vol 10 (3) ◽  
pp. 325
Author(s):  
Zhihao Wu ◽  
Baopeng Zhang ◽  
Tianchen Zhou ◽  
Yan Li ◽  
Jianping Fan
Keyword(s):  
Action Recognition ◽  
State Of The Art ◽  
Automatic Detection ◽  
Experimental Results ◽  
Practical Approach ◽  
Detection And Identification ◽  
Art Methods ◽  
Image Set ◽  
Social Images ◽  
Relationship Identification

In this paper, we developed a practical approach for automatic detection of discrimination actions from social images. Firstly, an image set is established, in which various discrimination actions and relations are manually labeled. To the best of our knowledge, this is the first work to create a dataset for discrimination action recognition and relationship identification. Secondly, a practical approach is developed to achieve automatic detection and identification of discrimination actions and relationships from social images. Thirdly, the task of relationship identification is seamlessly integrated with the task of discrimination action recognition into one single network called the Co-operative Visual Translation Embedding++ network (CVTransE++). We also compared our proposed method with numerous state-of-the-art methods, and our experimental results demonstrated that our proposed methods can significantly outperform state-of-the-art approaches.

scholarly journals Ensemble-Based Out-of-Distribution Detection

Electronics ◽  
2021 ◽  
Vol 10 (5) ◽  
pp. 567
Author(s):  
Donghun Yang ◽  
Kien Mai Mai Ngoc ◽  
Iksoo Shin ◽  
Kyong-Ha Lee ◽  
Myunggwon Hwang
Keyword(s):  
Detection Method ◽  
State Of The Art ◽  
Metric Learning ◽  
Feature Space ◽  
Confidence Score ◽  
Distance Metric Learning ◽  
Current State ◽  
Overall Performance ◽  
Deep Learning Model

To design an efficient deep learning model that can be used in the real-world, it is important to detect out-of-distribution (OOD) data well. Various studies have been conducted to solve the OOD problem. The current state-of-the-art approach uses a confidence score based on the Mahalanobis distance in a feature space. Although it outperformed the previous approaches, the results were sensitive to the quality of the trained model and the dataset complexity. Herein, we propose a novel OOD detection method that can train more efficient feature space for OOD detection. The proposed method uses an ensemble of the features trained using the softmax-based classifier and the network based on distance metric learning (DML). Through the complementary interaction of these two networks, the trained feature space has a more clumped distribution and can fit well on the Gaussian distribution by class. Therefore, OOD data can be efficiently detected by setting a threshold in the trained feature space. To evaluate the proposed method, we applied our method to various combinations of image datasets. The results show that the overall performance of the proposed approach is superior to those of other methods, including the state-of-the-art approach, on any combination of datasets.

Sign in / Sign up

Export Citation Format

Share Document